online labor marketsin this paper, between the two types of online labor markets, we focus on...

The IBIT ReportProviding Cutting-Edge Knowledge to Industry Leaders February 2013

A publication of the Institute for Business and Information Technology

Kevin Yili HongTemple University

Paul A. PavlouTemple University

Online Labor Markets:An Informal “Freelancer Economy”

© February 2013Institute for Business and Information TechnologyFox School of BusinessTemple University

Online Labor Markets: An Informal “Freelancer Economy”

Kevin Yili HongTemple University

Paul A. PavlouTemple University

4 The IBIT Report

ONLINE LABOR MARKETS

The IBIT ReportBruce FademEditor-in-chiefRetired VP and CIO, Wyeth

Paul PavlouEditorProfessor and Stauffer Senior Research FellowFox School of Business, Temple University

Laurel MillerManaging EditorDirector, Fox School of Business, Temple University

Board of Editors

Andrea AnaniaRetired VP and CIO, CIGNA

Ed BeaumontManagement Consultant

Michael BradshawVice President, Lockheed Martin

Jonathan A. BrassingtonFounding Partner and CEOLiquidHub Inc.

Richard CohenManaging Director, Deloitte

Larry DignanEditor-in-Chief, ZDnetSmartPlanet Editorial Director, TechRepublic

Craig ConwayPresidentConway Technology Consulting, Inc.

David KaufmanSenior VP and CIO, Aramark

Niraj PatelManaging Director, Witmer, LLC

Kent SeinfeldRetired CIO, Commerce Bank

Joseph SpagnolettiSenior VP & CIO, Campbell Soup Co.

© 2013 Institute for Business and Information Technology, Fox School of Business, Temple University, Philadelphia, PA 19122, USA. All rights reserved. ISSN 1938-1271.

The IBIT Report is a publication for the members of the Fox School’s Institute for Business and Information Technology. IBIT reports are written for industry and based on rigorous academic research and vendor neutral analysis. For additional reports, please visit our website at http://ibit.temple.edu.

No part of this publication may be reproduced, stored in a retrieval system or transmitted in any for or by any means, electronic, mechanical, photocopying, recording, scanning or otherwise, except as permitted under Sections 107 or 108 of the 1976 United States Copyright Act, without the prior written permission of the Publisher. Requests to the Publisher for permission should be addressed to Institute for Business and Information Technology, Fox School of Business, Temple University, 1810 N. 13th Street, Philadelphia, PA 19122, USA, 215.204.5642, or [email protected].

Disclaimer: The conclusions and statements of this report are solely the work of the authors. They do not represent the opinion of Temple University or the members of the Fox School’s Institute for Business and Informational Technology (IBIT).

Review process: IBIT reports are sourced, reviewed, and produced as follows. The managing editor oversees this process. The editor and editor-in-chief consult on topics and identify new sources of reports. The editor typically works with authors who are interested in writing reports and provides feedback on initial drafts. Completed reports are first screened by the editor-in-chief. After approval thereport is sent out for review to two members of the editorial board. The editor-in-chief assesses the completed reviews and provides further guidance to authors. Final reports are then professionally produced and made available to IBIT members.

5ibit.temple.edu

Foreword

IBIT Reports are sourced, reviewed, and produced by the editorial staff. The editor and editor-in-chief consult on topics and identify new sources of reports. The editor typically works with

authors who are interested in writing reports and provides feedback on initial drafts. Completed reports are first screened by the editor-in-chief. After his approval, the report is sent out for review by two members of the editorial board. The editor-in-chief assesses the completed reviews and provides further guidance to authors. Final reports are then professionally produced and made available to IBIT members.

Online labor markets have become significant global players linking companies with independent IT service providers. This IBIT Report Online Labor Markets: An Informal “Freelancer Economy” examines how firms like Freelancer allow the placement of request for quotes (RFQs) and the receipt of responses from service providers at very low search costs. The operating mechanics of the online marketplace are explained as well as the estimated economic benefits. The report then discusses the practical issues associated with operating in this environment including the selection of service pro-viders, monitoring of service providers, and managing language barriers. Online labor markets will continue to grow and anyone thinking of participating will benefit from the insights of this report.

Bruce FademEditor-in-Chief

January 28, 2013

6 The IBIT Report


“By changing the way work is done, online markets may lead to a new kind of economy centered on the individual.”

---Thomas W. Malone and Robert J. Laubacher, Harvard Business Review, 1998

IntroductionIn online labor markets, such as freelancer.com,

eLance.com, and Guru.com, companies can outsource various IT services, such as website development, graphical design, or creative writing to independent service professionals. The potential of these markets is widely touted (Malone and Laubacher 1998), and they have been expanding at an astounding pace since their inception. Forbes Magazine calls these online labor markets “an escape hatch for unemployed professionals” particularly during the economic crisis.

The Bureau of Labor Statistics estimates that more than 30 million people now work as independent professionals in the United States alone. Simply looking at the publicly available statistics for one of the major players – www.freelancer.com (Freelancer)- we can see that the scale of the “freelancer economy” is anything but trivial. These online labor markets serve as intermediaries for IT services (buyer companies post requests for quotes (RFQs) for services and providers place bids to win contracts from buyer companies) that help match companies with service providers across the globe at relatively low search costs.

Figure 1: Statistics from www.freelancer.com

(November 9, 2012)

7ibit.temple.edu

Stage 1. Before Project

Stage 2. Contracting Stage 3. Post-Project

Buyer post a project onto the marketplace (RFQ), with requirements and an estimated budget

Providers bid for services with a bid price, and buyer chooses a provider to work with her. A contract is now in effect.

The provider finished the project and send delivery. The provider and the buyer provide feedback to each other.

Figure 2 shows a typical RFQ page posted by a service buyer on a typical online labor market for services. And Figure 3 shows a typical project contracting process.

In this report, we discuss three aspects of online labor markets. First, we give an overview of the current state of major online marketplaces that offer intermediary services for buyer companies and IT professionals with a focus on one marketplace in particular – Freelancer. Second, we show some findings on the measurable economic benefits online

labor markets provide to companies that outsource IT services (measured with consumer surplus) and what factors contribute to higher buyer satisfaction. Third, we offer a novel perspective on the frictions that exist in the micro-level interactions between buyer companies and service providers given the global nature of these markets and how reputation signals can alleviate these frictions. Finally, we offer practical implications and recommendations for companies that use these marketplaces to outsource IT services to professionals across the globe.

Figure 2: Buyer’s Request for Quote (RFQ)

Figure 3: Project Process

8 The IBIT Report


Major Players and TypesMajor Players

Currently there are mainly two types of online labor marketplaces: reverse-auction type outsourcing marketplaces and contest marketplaces. Reverse-auction outsourcing marketplaces are the most common and comprehensive marketplaces, and in these markets, service providers start working after they are contracted by buyer companies. Contest marketplaces (such as innocentive.com, freecontest) mainly specialize in innovation contests that work in a rather different way than outsourcing marketplaces. In those markets, the contestants need to submit the product (such as a graphical design) before they are available to be chosen by the buyer company. In this paper, between the two types of online labor markets, we focus on outsourcing marketplaces. Table 1 offers an overview of the major players in this space. A comprehensive list of major players is offered in Appendix 2.

Table 1. Major Online Marketplaces for Outsourcing IT Services

Online labor marketplaces deal with different categories of projects. For Freelancer, as an example, there are mainly 10 categories of projects, including (1) Software Development, (2) Graphical Design, (3)

Content, (4) Data Entry, (5) Engineering and Science, (6) Product Sourcing, (7) Sales and Marketing, (8) Business and Accounting, (9) Translations and (10) Miscellaneous. The first four major categories account for more than 95% of all the projects. Below is a figure of the distribution of project types.

Freelancer

Our specific research site is Freelancer. Freelancer is an Australian based IT firm that acts as a global online labor marketplace that connects millions of service providers with buyer companies for outsourcing service projects. It began operating in 2004 and quickly evolved into a leading reverse auction type outsourcing marketplace for various services. Until November 2012, over 6 million service providers completed about 4 million service contracts ($660 million in transaction volume). Freelancer recently acquired vWorker (formally rentAcoder.com) in November 2012. Freelancer has been widely reported and covered by major media, such as Wall Street Journal, New York Times, the Economist Magazine and Forbes.

Figure 4: Composition of Project Types on Freelancer

9ibit.temple.edu

Marketplace Project Mechanisms

Online labor marketplaces often follow a scoring auction mechanism where buyer c o m p a n i e s ( o u t s o u r c e r s ) initiate posts to the

marketplace portal seeking bids from service providers who have related expertise. Along with this “request for quotes”, the buyer company provides a project description, required skills (such as java, php, python, etc.), proposed budget, and the auction end time. Registered service providers (IT professionals) read the project descriptions and decide whether to bid, and how much to bid. The majority of the projects on the marketplace follow an “open bids” format. Therefore, all service providers can see both the competing bids and the characteristics of the providers who submit those bids. The intermediary ends the auction by the date pre-specified by the buyer company (auction duration). The buyer company observes all bids by the service providers (including location, reputation, prior projects etc.) and makes a decision on (a) whether to award the contract to any provider who submitted a bid, and (b) whom to award the contract to.

Estimated Economic Benefits of Online Labor Markets

We focus on consumer surplus as a measure for the economic benefits online labor markets offer to buyer companies. Consumer surplus is an economic concept defined as the monetary gain obtained by consumers because they are able to purchase a product for a price that is less than the highest price

that they would be willing to pay. Consumer surplus and buyer satisfaction thus serve as key measures for the economic benefits of online labor markets. Therefore, they are important indicators of market success, from the perspective of the demand (buyer).

We analyzed a random sample of software development projects from Freelancer and estimated consumer surplus as a measure of the benefits from such markets for buyer companies. Our estimated consumer surplus is $136.9 per an average software development project. Given that the average bid price for a project in the sample is $160 and the average winning bid (contracted price) is $137.7, the level of consumer surplus ($136.9) is surprisingly high, indicating that on average, buyer companies saved more than 80% of their budget per outsourced project contracted on Freelancer, which is a notable economic effect.

We continue by looking at the factors that affect the satisfaction of buyer companies after the project is completed. The effects of the identified factors on buyer satisfaction are summarized in Table 2, and we explain these factors below.

Figure 5: Summary Statistics for Projects on Freelancer

10 The IBIT Report


Table 2. Factors that Affect Buyer Satisfaction (Hong, Pavlou and Chen 2011)

increases satisfaction because service providers from lower PPP countries are likely to work harder because they observe a higher valuation of the same monetary award relative to providers from higher PPP countries. This, in turn, increases the satisfaction and surplus of buyer companies in global online labor markets.

Language Barrier

Language barrier is a major issue on global online markets since effective communication relies on good command of a common language. Transaction partners from the same country usually can communicate requirements, ideas, and feedback better. Accurate usage of words can eliminate ambiguity, reduce communication costs, and avoid redundant work. Since English is the primary language in global online markets, if both parties fluently speak English would determine whether the service requirements could be effectively communicated. Therefore, the language barrier reduces the satisfaction of buyer companies in global online labor markets.

Repeat Transactions

Studies found that buyer companies tend to be more willing to select providers with whom they transacted before (Gefen and Carmel 2008) because

Global Labor Arbitrage

Global labor arbitrage is an economic phenomenon where, as a result of the removal of barriers to international trade, jobs move to nations where the cost of labor is relatively cheap. This phenomenon was observed by the New York Times (Roach 2004): “Under unrelenting pressure to cut costs, American companies are now replacing high-wage workers here with like quality, low-wage workers abroad. With new information technologies allowing products and now knowledge-based services to flow more easily cross borders, global labor arbitrage is likely to be an enduring feature of the economy.” Global labor arbitrage affects buyers’ decisions since companies are more likely to choose a service provider from poorer country to take advantage of low labor cost, thus increasing their consumer surplus. This is partly because the valuation of $1 dollar varies across countries. The utility a service provider can derive from a certain monetary amount also depends on the purchasing power of that amount in the provider’s home country. Gefen and Carmel (2008) found that companies tend to choose providers from poorer countries; their argument is that for the same amount of monetary award, companies would expect service providers from poorer countries to work harder. We argue that the purchasing power of service providers

1 PPP is defined as purchasing power adjusted GDP per capita in a country. It is a measure for global labor arbitrage.

11ibit.temple.edu

they are perceived to have lower transaction costs. From the buyer’s perspective, if the buyer company is satisfied with a provider’s performance, the company is likely to continue using the service provider. Subsequent (dyadic) transactions are likely to be smoother and faster as the provider knows what the buyer company needs. Therefore, repeat dyadic transactions (versus first-time or one-time transactions) are positively associated with the satisfaction of buyer companies.

Buyer Experience

As buyer companies accumulate experience in the labor market, they learn to avoid losses from opportunistic or low quality providers, and they

have more realistic expectations of provider quality. First, as Nicholas Carr (2003) argued, in markets with asymmetric information, bid evaluation is costly. Therefore, when multiple bids are available, finding a service provider that would offer the highest satisfaction is a formidable task. Experienced buyer companies are more comfortable in awarding contracts to low bidders with fair quality. Second, experienced buyer companies that have dealt with many service providers in the past are more likely to incur lower evaluation costs, and they are more effective and efficient in evaluating all bids to make an informed decision. Third, experienced buyer companies are more likely to spot problematic service providers and avoid adverse selection (picking a low quality provider with but average price). These reasons suggest that more experienced buyer companies would have higher satisfaction.

Transaction Interactions between Service Providers and Buyer Companies

Having looked at the key factors that contribute to buyer satisfaction above, we take a closer look at the micro level interactions between service providers and buyer companies, especially how these factors will affect the two important aspects of decisions in the online labor markets, pricing of services by providers and the selection of service providers by buyer companies. After the buyer companies post jobs on the marketplace, providers who are interested first make a decision of “how much to bid” (pricing decision) to offer the job. Since there are usually multiple providers bidding on a job, the buyer company needs to make the decision of “whom to award the contract to” (selection decision).

For a buyer company, different service providers are associated with different levels of frictions due to the geographical dispersion and innate complexity of IT services. For example, the buyer company and the service provider may speak different languages, and they are likely to have a communication barrier. Also the buyer company and the service provider may be located in different time zones, which might give rise to the challenge of effective communication during regular business hours. Beyond language and time zone differences, cultural differences may pose additional frictions. Therefore, for different service providers, the same buyer company may be associated with different levels of transaction frictions. We summarize these issues as “global frictions.”

IT services are specific, complex (Snir and Hitt 2003) and non-contractible (Brynjolfsson and Smith 2000). These attributes suggest that IT services are for a specific purpose, they are hard to contract on because they are innately complicated, leading to

Therefore, when multiple bids are available, finding a service provider that would offer the highest satisfaction

is a formidable task.

12 The IBIT Report


effect of global frictions on the bid price of service providers and their selection by buyer companies. Buyer companies select a service provider to maximize their utility, therefore, they are less likely to choose a service provider who speaks a different language or lives in a different time zone; instead, they would prefer a service provider who they have transacted before.

Different service providers offer different prices for the buyer based on the degree of uncertainty buyers face when transacting with them. It is well established that imperfectly informed buyers

discount their prices (Milgrom and Weber 1982). We summarize the major factors that constitute global frictions (1) lack of familiarity (whether the buyer and provider have transacted before), (2) language difference, (3) time-zone difference, and (4) culture difference. We show below the world’s language, time-zone and cultural map. Table 3 shows our results on how these factors affect the service provider’s pricing and their selection by buyer companies based on their bid price.

incomplete contracts. Plus it is difficult to monitor the service provider’s efforts due to geographical and temporal separation, resulting in potential for moral hazard (the service provider’s level of effort may change, to the detriment of the buyer company after a transaction takes place). In most online labor markets, a reputation system is in place to mitigate these issues. Not only do service providers observe their competitors’ bid prices, they observe their competitor’s characteristics, such as their reputation scores as well. Therefore, when service providers signal their quality to each other, the reputation signals may be picked up by other service providers to make pricing decisions. We summarize the second factor as “reputation signals.”

In sum, we identified two key aspects that affect the selection of service providers and the pricing of their services: global frictions and reputation signals.

Global Frictions

We take the lens of uncertainty to understand the

We identified two key aspects that affect the selection of service providers and the pricing of their services: global

frictions and reputation signals.

13ibit.temple.edu

Figure 6: World Language Map

Figure 7: World Time-zone Map (International Telecommunication Union)

Figure 8: Cultural Difference (World Value Survey)

14 The IBIT Report


Reputation Signals

Reputation signals are quality indicators of service providers. Freelancer has established a reputation system, which mainly include average feedback rating, project experience and gold member status. Our research indicates that providers with higher feedback rating, more project experience and a gold member status are seen to be more capable and reliable by buyer companies. From the provider’s perspective, one insight from the utility theory is that service providers will bid rationally within the bounds of their information. To achieve the same level of utility, high quality service providers rationally require higher compensation than low quality service providers (incentive compatibility). In open bids marketplace, as reputation signals are available to both buyer companies and all service providers, high quality service providers can leverage these signals to

bid a premium and still win the contract. We call this competitive pricing. Competitive pricing is the case that each service provider observes other providers’ reputation signals relative to their own, to decide on his own bid price.

We found the effect of these two key factors (global frictions and reputation signals) to co-exist in determining the providers’ bidding price and the buyer company’s selection of service providers. The directions of effects are shown in Table 3. We found that all four types of global frictions were seen negatively by buyer companies; however, only lack of familiarity and language difference would increase a provider’s bid price. The results indicate that providers are very sensitive to lack of familiarity and language difference. Beyond global frictions, providers with better reputation signals (average feedback rating, project experience and gold member status) have a higher probability of being awarded the contract by the buyer company, and they tend to bid a higher price as well.

Table 3. Effect of Global Frictions and Reputation Signals (From Hong and Pavlou 2012)

15ibit.temple.edu

Discussion and Political Implications

In this section we discuss the implications of this study on global online labor markets, and we offer practical recommendations on two main aspects. The first aspect focuses on current issues with global online labor markets, such as monitoring and global frictions. The second aspect focuses on how to design better marketplaces to prevent a market of lemon, and what type of projects buyer companies should consider using online labor.

Monitoring of Service Providers

IT services on online labor markets are unique. Unique products/services are difficult to contract, and incomplete contracts lead to the potential for post-transaction conflict. The source of conflict is asymmetric information (a situation in which one party in a transaction has more information than another) between the buyer company and the service provider because the provider’s actions cannot be perfectly observed and perfectly contracted upon before the contract is signed (Hölmstrom 1979). To make things even worse, geographical separation renders a natural remedy to conflict – performance monitoring - virtually impossible. Contracts in markets with asymmetric information are based on reputation and trust. When buyer companies and service providers are geographically dispersed, physical performance monitoring becomes difficult, allowing service providers to “shirk”, potentially resulting in moral hazard and conflict (WSJ 2008). Incomplete contracts and moral hazard faced by markets with asymmetric information may lead to the problem of service providers not performing according to the

expectations of buyer companies. Therefore, we have the following recommendations for buyer companies, service providers, and the marketplace.

First, since the value and sustainability of online labor markets depend to a large extent on “job supply”, it is important for these markets to increase customer benefits with superior designs. As we empirically found that experienced buyer companies enjoy higher surplus, the intermediary should educate new buyer companies how to navigate its marketplace so they do not suffer from low surplus that may cause them to

exit the market. Also, we found that the language barrier reduces consumer surplus; it is important for signals about the provider’s language proficiency (English or other languages)

to be visible on the marketplace’s platform.

Second, for buyer companies, we found that with the same amount of money, on average, service providers from poorer countries will work harder than those from richer countries. The analyses show that the positive effect of the English language on the buyer company’s satisfaction is attenuated by the increase in PPP (e.g., India providers will provide a higher consumer surplus than US/UK providers). Therefore, countries like India will be ideal for providing satisfactory services, while providers from

We found that with the same amount of money, on average, service providers from poorer countries will work harder

than those from richer countries.

16 The IBIT Report


low PPP non-English speaking (China) or English speaking high PPP countries (US) will be the second tier, and providers from non-English speaking high PPP countries are the least favorable in terms of buyer satisfaction.

Global Frictions

In this study, we found the “world is flat” claim to be challenged by various frictions that threaten online global markets from being level playing fields. We show that service providers are sensitive to the uncertainty caused by intangible online frictions. Given that the sensitivity is high, service providers charge a price premium for frictional costs due to lack of prior transactions and a language differences in order to shield themselves from uncertainty; albeit service providers are shown to have low sensitivity to time zone differences and cultural differences. One reason for no effect of time zone differences is that many of the service providers are freelancers, and it is their routine to work after hours, so that time zone difference may not be a big deal for them.

How to Prevent a “Market of Lemons”

A market will become a “market of lemons” when only the low-quality products (“lemons”) will stay in the market as the high-quality products exit the markets since they cannot distinguish their high quality due to information asymmetry. For example, in the market for used cars, buyers cannot perfectly ascertain the quality of a used car, and they may thus only be willing to pay an average price of all similar used cars (irrespective whether it is a low

or high quality car). In this situation, owners of high-quality cars cannot obtain premium prices as returns to their superior quality and will therefore simply exit the market. Online labor markets are likely to become “market of lemons” if high quality service providers cannot signify their high quality to obtain a price premium, they will likely exit the market, instead of accepting an average price that does not distinguish between high and low-quality service providers. In this study, we find evidence for the effect of reputation signals (i.e., average feedback rating, past transactions, and gold member status) to ensure that high quality service providers do reap a price premium compared to low quality providers. This indicates that high quality service providers do charge a price premium for their good reputation. It is important for the marketplace to ensure the reputation signals are visible, clear and

credible. For example, the marketplace needs to make sure that the reputation scores are placed at spots on the webpage that are easily located by buyer companies.

Project Size Matters!

Our results highlight both the opportunities and the challenges that might be expected transacting services

through online markets. Eli Snir and Lorin Hitt observed: “As many types of services produce digital products and do not require large-scale production capital that drives the formation of large firms, the

In this study, we find evidence for the effect of reputation signals to ensure that high quality service providers do reap a price premium compared to low

quality providers.

service industries should be even more amenable to increased outsourcing and the erosion of firm boundaries.” Indeed, we have seen that in the IT domain, technologies that promote interoperability (e.g., object-oriented techniques) and best management practices (e.g., the practice of subdividing larger projects into smaller stages) also favor the outsourcing of small-scale projects. However, evidence suggests that large projects tend to fail through in higher probability. And we recommend to buyer companies to be cautious in crowdsourcing large IT projects (>$100,000) via online global labor markets, at least for now. If large projects are to be outsourced, proper monitoring, effective communications and immediate feedback are important factors leading to a successful IT project. Our results show that for larger value projects, buyer companies are less likely to be satisfied with the results and less likely to gain a significant economic benefit. However, for small firms (especially start-ups or start-up units in large companies), global online labor markets may provide an ideal workforce at low price; and for such companies, we recommend using these markets, to gain a competitive edge at a fraction of the regular price through traditional (local) service providers.

Conclusion Online global labor markets serve as intermediaries

that help companies to find service providers (e.g., software vendors, programmers, and graphical designers) at a low cost. Although this industry is at a developing stage, it has already created a substantial social welfare and provided a lot of job opportunities and income to IT professionals. In this report, we presented results on the current industry development, social benefits of such global labor markets, and practical issues pertaining to online labor contracting that may affect service providers and buyer companies. We believe online labor markets will continue to grow as we are heading toward a “freelancer economy”.

Figure 9: Project Size and Contract Rate (From Snir and Hitt 2003)

18 The IBIT Report


ReferencesBanker, R., and Hwang, I. 2008. “Importance of Measures of Past Performance: Empirical Evidence on

Quality of E-Service Providers,” Contemporary Accounting Research (25:2), pp 307-337.

Carr, S. 2003. “Note on Online Auctions with Costly Bid Evaluation,” Management Science (49:11), pp 1521-1528.

Gefen, D., and Carmel, E. 2008. “Is the World Really Flat? A Look at Offshoring at an Online Programming Marketplace,” MIS Quarterly (32:2), pp 367-384.

Hölmstrom, B. 1979. “Moral Hazard and Observability,” The Bell Journal of Economics (10:1), pp 74-91.

Hong, Y., and Pavlou, P.A., 2012. “An Empirical Investigation on Provider Pricing in Online Crowdsourcing Markets for IT Services”, Proceedings of 2013 International Conference on Information Systems, Orlando, FL.

Hong, Y., Pavlou, P.A. and Chen, P-y, 2011. “Quality-adjusted Consumer Surplus in Markets with Asymmetric Information”. The Workshop on Information Systems and Economics, Shanghai, China

Malone, T.W., and Laubacher, R.J. 1998. “The Dawn of the E-Lance Economy,” Harvard Business Review (76).

Milgrom, P.R., and Weber, R.J. 1982. “A Theory of Auctions and Competitive Bidding,” Econometrica: Journal of the Econometric Society, pp 1089-1122.

Roach, S.S. 2004. “More Jobs, Worse Work,” the New York Times), pp http://www.nytimes.com/2004/2007/2022/opinion/more-jobs-worse-work.html.

Snir, E., and Hitt, L. 2003. “Costly Bidding in Online Markets for It Services,” Management Science (49:11), pp 1504-1520.

WSJ. 2008. “The War on Returns,” in: Wall Street Journal. http://online.wsj.com/article/SB121020824820975641.html Published May 8. Reporter: Christopher Lawton.

19ibit.temple.edu

Appendix 1. Integrated Dataset and Sample

Our samples come from two archival sources. First, transaction data consisting of a sample of 159,748 RFPs from 38,315 buyers, that took place between February 4, 2004 and September 24, 2010, was obtained from the online market with MySQL query. The outsourcing services were across different project categories. Second, we obtained the PPP adjusted GDP per capita indices and the official language data from the CIA World Factbook1, which is widely used in prior research (Gefen and Carmel 2008). We also obtained PPP data from the International Monetary Fund2 and the World Bank3 to amend the CIA World Factbook data because PPP data were not available in the CIA World Factbook. We compared indices from the three data sources to ensure consistency. We combined the PPP-adjusted GDP data and English-speaking country data with the first dataset. Third, we drew on a dataset constructed by Mayer and Zignago (2005) for their study of international trade. This dataset includes three official languages for 246 countries and independent regions, as well as an indicator of shared major languages (a binary

1 CIA World Factbook: http://www.cia.gov/cia/publications/factbook/geos/ae.html/. 2 International Monetary Fund, World Economic Outlook Database. 2010 data.3 World Development Indicators database, World Bank - 29 September 2010

indicator reflecting the existence of at least one language, spoken in both countries, by at least 7% of the respective populations). We chose to employ a binary indicator of at least one shared language. Though a binary indicator is likely to be noisier, we opted for this out of a desire for simplicity. Wherever possible, additional data on language or religion was supplemented from the CIA World Factbook to replace missing values. The shared language data is also merged to the main transaction data set. Fourth, we draw on the latest report4 by International Telecommunication Union (ITU) on world legal time to obtain our raw data on time zone of the countries in our transactional data set. We calculate the absolute time zone difference based on the raw data for the provider-buyer countries pairs. Fifth, we draw on a data set composed by World Value Survey on cultural differences. We rely on the measure used in prior research, which captures the Euclidean distance between pairs of countries with regard to two dimensions of culture: secular/rational values (the extent to which a society emphasizes traditional as opposed to secular values) and the survival versus self-expression orientations (the extent to which a society emphasizes values related to survival as opposed to self-expression). These two dimensions explain 70% of the variations in the features of different cultures.

4 The report can be downloaded from here: http://www.itu.int/dms_pub/itu-t/opb/sp/T-SP-LT.1-2011-PDF-E.pdf

20 The IBIT Report


Appendix 2. A Comprehensive List of Online Labor MarketplacesMarketplace Website Type

I can freelance http://www.icanfreelance.com/ ProjectiFreelance www.ifreelance.com ProjectKi Work www.ki-work.com ProjectNet4manpower www.net4manpower.com ProjectoDesk www.odesk.com ProjectPeople per hour www.peopleperhour.com ProjectProcess Outsourcing www.contractedwork.com ProjectPROJECT FOR HIRE www.project4hire.com ProjectProjectSpring.com www.projectspring.com ProjectSerebra connect www.serebraconnect.com ProjectTrada www.trada.com ProjectUnited states freelancers http://www.unitedstatesfreelancers.com ProjectuTest http://www.utest.com ProjectVerbum soft http://directfreelance.com/default.aspx ProjectVois www.vois.com ProjectWork Outsource www.workoutsource.net Project99designs http://99designs.com/ Graphic DesignCreateAd www.creatad.com Graphic DesignCrowdspirit www.crowdspirit.com Graphic DesignCrowdSpring www.crowdspring.com/ Graphic DesignDesign Outpost www.designoutpost.com Graphic DesignDesignBay www.designbay.com Graphic DesignDesignContest.net www.designcontest.net Graphic DesignDesignQuote.net www.designquote.net Graphic DesignGeniusRocket www.geniusrocket.com Graphic DesignGFXContests www.gfxcontests.com Graphic DesignGraphic Competitions www.graphiccompetitions.com Graphic DesignLogoTournament http://logotournament.com/ Graphic DesignProva Advertising http://www.prova.fm/advertising/ Graphic DesignFreelance CGI programmers www.cgilance.com Programmer EUFREELANCE www.eufreelance.com Programmer Freelance Web Programming www.freelancewebprogramming.com Programmer GetACoder.com www.getacoder.com Programmer Lime exchange www.limeexchange.com Programmer Programmermeetdesigner www.programmermeetdesigner.com Programmer Programming bids http://www.programmingbids.com/ Programmer Rent A Coder www.rentacoder.com Programmer Scriptlance www.scriptlance.com Programmer TopCoder www.topcoder.com Programmer

21ibit.temple.edu

About the authorsKevin Yili Hong is a doctoral candidate and an instructor of Management Information Systems (MIS) at the Fox School of Business, Temple University. He graduated magna cum laude from Beijing Foreign Studies University (China) with a B.S. in Management and a B.A. in English Literature. His research focuses on online marketplaces, e-commerce and economics of IS. He has published in MIS Quarterly (MISQ), Journal of Global Information Management (JGIM), Workshop on Infor-mation Systems and Economics (WISE), International Conference on Information Systems (ICIS), Hawaii International Conference on Systems Sciences (HICSS), the Academy of Management Conference (AoM), among others. He has won best paper award at ICIS, AoM and AMCIS. Yili serves as a reviewer for journals such as Man-agement Science, MISQ, ISR, JAIS, I&M, etc. Prior to pursuing a Ph.D. in MIS, he has worked in the largest investment bank in China (China International Capital Corporation). Yili consults several leading Internet-based companies such as www.freelancer.com, and www.ookong.com. And he was an interpreter for the President of International Olympic Committee (IOC) Jacques Rogge in the 29th Olympic Games in 2008.

Contact information:Yili HongFox School of Business, Temple University201F, 1810 N. 13th Street. Speakman Hall Philadelphia, PA 19122, [email protected], (215) 327-7501

22 The IBIT Report


About the authorsPaul A. Pavlou is a Professor of Management Information Systems, Marketing, and Management, director of the Ph.D. program and a Stauffer Senior Research Fellow at the Fox School of Business and Management at Temple University. He received his Ph.D. from the University of Southern California in 2004. His research focuses on e-commerce, online auctions, information systems strategy, information economics, research methods, and NeuroIS. His research has appeared in MISQ, ISR, JMIS, JAIS, JAMS, CACM, and Deci-sion Sciences, among others. His work has been cited over 1,300 times by the Social Science Citation Index of the Institute of Scientific Information, and over 4,500 times by Google Scholar. Paul won several Best Paper awards for his research, including the ISR Best Pa-per award in 2007, the 2006 IS Publication of the Year award, the Top 5 Papers award in Decision Sciences in 2006, the Runner-Up to the Best Paperaward of the

2005 Academy of Management Conference, the Best Doctoral Dissertation award of the 2004 International Conference on Information Systems (ICIS), the Best Interactive Paper award of the 2002 Academy of Man-agement Conference, and the Best Student Paper award of the 2001 Academy of Management Conference (OCIS Division). Paul also won several Reviewer awards, including the 2009 Management Science Meritorious service award, the ‘Best Reviewer’ award of the 2005 Academy of Management Conference, and the 2003 MIS Quarterly ‘Reviewer of the Year’ award. Paul sits on the Editorial Boards of MISQ, IJEC, and ECRA. He is currently a guest Senior Editor of a Spe-cial Issue of MISQ on “Novel Perspectives of Trust in Information Systems” published in June of 2010, and as a guest Senior Editor of a Special Issue of JMIS on ‘Trust in Online Environments’ published in 2008.

Contact information:Paul A. Pavlou

Fox School of Business, Temple University201D, 1810 N. 13th Street. Speakman Hall

Philadelphia, PA 19122, [email protected], (215) 204 - 3583

23ibit.temple.edu

Fox School of Business and Management

Established in 1918, the Fox School of Business has a distinguished tradition of preparing business leaders, professionals and entrepreneurs for successful careers. Today, it is the largest, most comprehensive business school in the Greater Philadelphia region, and among the largest in the world with over 6,500 students, 180 full-time faculty and more than 59,000 alumni.

In 2013, The Fox MBA in Information Technology Management ranked top 25 and in 2012 and 2013 and the Bachelor of Business Administration (BBA) in MIS ranked top 20 in the nation by U.S. News & World Report. The Association for Information Systems ranked Fox School’s MIS faculty number 3 worldwide for research in 2011 and number 1 worldwide in 2010. In 2009, TechRepublic ranked the Fox BBA in MIS in the top 10 U.S. programs for information technology. In 2008, The Chronicle of Higher Education, ranked Fox School MIS faculty in the top 10 for research productivity.

Institute for Business and Information Technology

The Institute for Business and Information Technology (IBIT) provides cutting-edge knowledge and valuable connections to sustain excellence in information technology. IBIT integrates industry perspectives with academic research expertise to create forums for generating and exchanging best practices. IBIT membership offers participating firms the opportunity to leverage our knowledge, human capital, relationships, and established network.

SYMPOSIUMS AND CONFERENCESExclusive highly interactive forums of noted practitioners and academics addressing current topics.

THE IBIT REPORTRigorous vendor neutral research that provides actionable knowledge to industry.

PROJECTSIBIT faculty and students work with member firms on joint business technology projects.

IT AWARDSThe IT Innovator, Leader, and Distinguished Alumni awards are presented to industry leaders at the annual IT Awards Reception.

DISTINGUISHED SPEAKER SERIESFeatures talks by leading professionals on critical business technology topics.

INDUSTRY INTERNSHIPSProject-based internship model provides firms and students with a structured light weight opportunity to work on projects.

SCHOLARSHIPS AND FELLOWSHIPSRecognition for exceptional students as well as faculty.

EXECUTIVE-IN-RESIDENCEThe Executive-in-Residence program facilitates interaction between students, faculty and industry leaders.

ADVISORY BOARDThe Fox IT advisory board, chaired by Bruce Fadem, retired VP and CIO, Wyeth, includes senior executives from Lockheed Martin, Merck, LiquidHub, Deloitte, AstraZeneca, ZDNet, Pfizer, Aramark, Campbell Soup, Electronic Ink, and others.

The IBIT Report

The Fox School’s Institute for Business and Information Technology (IBIT) regularly pub-lishes The IBIT Report for its members. IBIT reports are based on rigorous vendor neutral academic research and are written to provide actionable knowledge to industry. Each report focuses on an important cutting edge topic that is of interest to our members.

For additional information, contact:Institute for Busines and Information TechnologyFox School of BusinessTemple University210 Speakman Hall (006-00)1810 N. 13th StreetPhiladelphia, PA 19122

[email protected]

online labor marketsin this paper, between the two types of online labor markets, we focus on...

Documents